#​507 — May 24, 2023

Web Version
Together with  pgAnalyze

Postgres Weekly

Building High-Perf Functions in Rust on Amazon RDS — Amazon RDS for PostgreSQL now supports trusted PL/Rust, opening up a new opportunity for building stored procedures in Rust for Postgres on the popular cloud provider. This post walks through a complete example. We expect to see some interesting use cases spring up from this!

Jonathan Katz (AWS)

The Draft Postgres 16 Release Notes — We’re still several months away from a final Postgres 16 release, but Bruce Momjian has been hard at work and finished the first draft of the eventual Postgres 16 release notes. “The release note feature count is 199, which is 7% higher than the average count of the previous five releases.”

PostgreSQL Documentation

Postgres Performance at Any Scale — Deliver consistent database performance and availability through intelligent tuning advisors and continuous database profiling. Take control of your Postgres workload and query plans and empower all development teams to fix slow queries. Try pganalyze today.

pganalyze sponsor

Tags and Arrays: A Purrrrfect Combination? — One for both cat and tag lovers. If you need to model a tagging system in your database, there are several approaches. Paul reviews and benchmarks a relational many-to-many table approach as well as using Postgres’s native array support while suggesting why you might choose either option.

Paul Ramsey

IN BRIEF:

  • Timescale, best known for their Postgres-based TimescaleDB time series extension, have sort of rebranded: "We used to be a database company with a cloud product. Now we are a cloud company with a database product." TimescaleDB remains open source, happily.

  • A look at using AI to help choose a shard key when working with Citus.

  • 📕 Ruby on Rails developer? High Performance PostgreSQL on Rails is an in-development book (to be published by Pragmatic Bookshelf) you can register an interest in.

Tuning Postgres and the New Insert Benchmark — If you’ve been using Postgres a while it’ll be no surprise that tuning your config settings can have an outsized effect on performance, but few folks seem to (publicly) benchmark as wide a variety of configs as this.

Mark Callaghan

Visualizing IoT Data at Scale with Hopara and TimescaleDB — Yes, that Michael Stonebraker, co-founder of the 'post-Ingres' effort that formed the Postgres we know today. How many other people can start a blog post with "I have been involved in building DBMSs for 50 years"?

Michael Stonebraker

Five Ways to Get Table Creation Information — When was a table created? The author covers five(!) places where you can get such information about Postgres objects.

Greg Sabino Mullane

Migrating Generated Columns to Postgres using AWS Database Migration Service — From databases like SQL Server or Oracle, say.

Karimi and Gopalappa (AWS)

Code and Tools

Postgres Query Lock Explainer — As the author says: “it’s kind of like EXPLAIN but it tells you about what locks would be required by the query.” Written in Python.

Adam Tal